AI Engineering for Software Engineers
How do you integrate LLMs into production as a Software Engineer? Master the full chain: APIs, RAG, agents and the Model Context Protocol.
What you'll learn
Integrate LLMs into production
Master Transformer architecture, LLM APIs (function calling, structured output) and engineering-grade prompting for robust integrations.
Design RAG systems
Build high-performing RAG pipelines: ingestion, chunking, vector search, reranking, and evaluation with the RAGAS framework.
Build AI agents
Implement the ReAct and Plan-and-Execute patterns, and connect your agents to tools via the Model Context Protocol (MCP).
The detailed programme
Nine progressive modules, from the theoretical foundations of LLMs to building autonomous AI agents connected via MCP.
LLM Foundations
3h30
The essential basics: neural networks, Transformer architecture, the attention mechanism and sampling strategies.
Objectives
- Understand neural networks, how they're trained, and the shift from RNNs to Transformers
- Master the attention mechanism (Query, Key, Value), embeddings and tokenization
- Understand autoregressive generation and sampling controls (Temperature, Top-K, Top-P)
What's covered
- Neural networks, training, and the shift from RNNs to Transformers
- Transformer architecture: attention mechanism (Query, Key, Value), embeddings, tokenization
- Autoregressive generation and sampling controls (Temperature, Top-K, Top-P)
- Pretraining, fine-tuning and alignment (RLHF, DPO)
Your experts
Training investment
1 500,00 €
per participant
9 000,00 €
per session
27
People trained (2026)
92%
Satisfaction (2026) · 12 responses
44%
Response rate (2026)
Fund your training through an OPCO
As a Qualiopi-certified training provider, the courses we offer can be funded through an OPCO (the French vocational-training funding body). Find which OPCO you depend on here.
Let's discuss your project
Personalised quotes, tailor-made formats, OPCO funding options: we answer all your questions.
Request a quote